I am an educational psychologist (Ph.D. University of Illinois, 1983) with expertise in cognitive psychology and psychometrics. Since 2004 I have been employed at the National Center for Education Statistics. From 1999 to 2004 I was employed at the Ball Foundation in Glen Elyn, Illinois
In this study, the relationships between simple learning by accretion and various cognitive abili... more In this study, the relationships between simple learning by accretion and various cognitive ability variables were explored. Computerized tests of five sources of individual differences were administered to a sample of 714 Air Force recruits, along with a trigram-English word paired-associate task, which was presented as a foreign language vocabulary learning task. Subjects were assigned at random to one of three groups: control, semantic elaboration, or interactive imagery. Subjects in the semantic elaboration group were instructed to generate sentences to link the trigram and word in a memorable way. Subjects in the interactive imagery group were given the additional instruction of visualizing the generated sentence. Trigrams (CVCs) varied in meaningfulness across the two lists of eight pairs in the task. Results showed that meaningfulness and strategy had the expected main effects on learning and that strategy interacted with verbal knowledge in initial learning so that learners with more knowledge benefitted more than learners with less knowledge edge. Regression analyses showed that a representative measure from each proposed source made a significant unique contribution to the explained variance in paired-associate learning.
ABSTRACT Includes bibliographical references (leaves 30-33) Supported in part by the National Ins... more ABSTRACT Includes bibliographical references (leaves 30-33) Supported in part by the National Institute of Education contract no. NIE-C-400-76-0016
Teachers College Record: The Voice of Scholarship in Education, 2007
Large-scale assessments designed to serve as indicators of academic progress in a social context ... more Large-scale assessments designed to serve as indicators of academic progress in a social context provide invaluable information about the condition of education in America. This unique class of assessments serves as a common yardstick by which the educational progress in states, jurisdictions, and other countries can be compared. Because these assessments serve as monitors across a wide variety of curricula, content standards, and instructional practices, they are uniquely designed and well suited for their task. The focus of this chapter is to define what policymakers need to know to be proficient in this kind of large-scale indicator assessment literacy. What does a large-scale indicator assessment bring to the table? As Resnick (1999) points out,
Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 1997
We plan to demonstrate a PC-based driving simulator developed by the US Air Force for assessing s... more We plan to demonstrate a PC-based driving simulator developed by the US Air Force for assessing situation awareness (SA). This simulator is a flexible tool that allows researchers to assess drivers' SA using a variety of different probes. The simulator allows assessment of explicit knowledge used in driving, as is done in a number of SA assessment tools. However, unlike other SA assessment tools, the simulator also allows assessment of implicit, automatized knowledge.
Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 1998
Because of the similarity of the AVO console to a desktop flight simulator, a preliminary idea of... more Because of the similarity of the AVO console to a desktop flight simulator, a preliminary idea of what aptitude factors would predict success in AVO training might be obtained through a validation study using a desktop flight simulator as the criterion task. In this study, 171 young men and women learned to fly a simulated light aircraft. The participants also took a comprehensive 8-hour battery of computer-based tests. Results indicated that learning was predicted well by general cognitive ability (g)/ working memory (WM) and multilimb coordination. When structural equation modeling was used to distinguish declarative and procedural learning, declarative learning was found to be strongly dependent on g/WM, which is consistent with previous research. That g/WM did not correlate with procedural learning explains why correlations of this ability with learning diminished as instruction moved from a declarative to procedural emphasis. Only multilimb coordination ability was strongly rel...
This research examined the dimensionality and the correlates of self-reported cognitive failures.... more This research examined the dimensionality and the correlates of self-reported cognitive failures. The first goal was to determine what factors, in addition to a general one, are needed to explain self-reported cognitive failures. To explore this issue, both Rasch measurement and confirmatory factor analysis were employed. The second goal was to determine if cognitive failures might be predicted with personality factors, general cognitive ability, and the need for cognition. A sample of 552 USAF airmen responded to the Broadbent Cognitive Failures Questionnaire (CFQ), a Big-Five personality inventory, the Reasoning Test, the Speeded Cognitive Ability Test, and the Need for Cognition survey. Both Rasch modeling and confirmatory factor analysis indicated that a single factor dominated CFQ responses. Regression analysis showed that CFQ responses were predicted well by personality factors (R = .60).
In this study, the relationships between simple learning by accretion and various cognitive abili... more In this study, the relationships between simple learning by accretion and various cognitive ability variables were explored. Computerized tests of five sources of individual differences were administered to a sample of 714 Air Force recruits, along with a trigram-English word paired-associate task, which was presented as a foreign language vocabulary learning task. Subjects were assigned at random to one of three groups: control, semantic elaboration, or interactive imagery. Subjects in the semantic elaboration group were instructed to generate sentences to link the trigram and word in a memorable way. Subjects in the interactive imagery group were given the additional instruction of visualizing the generated sentence. Trigrams (CVCs) varied in meaningfulness across the two lists of eight pairs in the task. Results showed that meaningfulness and strategy had the expected main effects on learning and that strategy interacted with verbal knowledge in initial learning so that learners with more knowledge benefitted more than learners with less knowledge edge. Regression analyses showed that a representative measure from each proposed source made a significant unique contribution to the explained variance in paired-associate learning.
ABSTRACT Includes bibliographical references (leaves 30-33) Supported in part by the National Ins... more ABSTRACT Includes bibliographical references (leaves 30-33) Supported in part by the National Institute of Education contract no. NIE-C-400-76-0016
Teachers College Record: The Voice of Scholarship in Education, 2007
Large-scale assessments designed to serve as indicators of academic progress in a social context ... more Large-scale assessments designed to serve as indicators of academic progress in a social context provide invaluable information about the condition of education in America. This unique class of assessments serves as a common yardstick by which the educational progress in states, jurisdictions, and other countries can be compared. Because these assessments serve as monitors across a wide variety of curricula, content standards, and instructional practices, they are uniquely designed and well suited for their task. The focus of this chapter is to define what policymakers need to know to be proficient in this kind of large-scale indicator assessment literacy. What does a large-scale indicator assessment bring to the table? As Resnick (1999) points out,
Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 1997
We plan to demonstrate a PC-based driving simulator developed by the US Air Force for assessing s... more We plan to demonstrate a PC-based driving simulator developed by the US Air Force for assessing situation awareness (SA). This simulator is a flexible tool that allows researchers to assess drivers' SA using a variety of different probes. The simulator allows assessment of explicit knowledge used in driving, as is done in a number of SA assessment tools. However, unlike other SA assessment tools, the simulator also allows assessment of implicit, automatized knowledge.
Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 1998
Because of the similarity of the AVO console to a desktop flight simulator, a preliminary idea of... more Because of the similarity of the AVO console to a desktop flight simulator, a preliminary idea of what aptitude factors would predict success in AVO training might be obtained through a validation study using a desktop flight simulator as the criterion task. In this study, 171 young men and women learned to fly a simulated light aircraft. The participants also took a comprehensive 8-hour battery of computer-based tests. Results indicated that learning was predicted well by general cognitive ability (g)/ working memory (WM) and multilimb coordination. When structural equation modeling was used to distinguish declarative and procedural learning, declarative learning was found to be strongly dependent on g/WM, which is consistent with previous research. That g/WM did not correlate with procedural learning explains why correlations of this ability with learning diminished as instruction moved from a declarative to procedural emphasis. Only multilimb coordination ability was strongly rel...
This research examined the dimensionality and the correlates of self-reported cognitive failures.... more This research examined the dimensionality and the correlates of self-reported cognitive failures. The first goal was to determine what factors, in addition to a general one, are needed to explain self-reported cognitive failures. To explore this issue, both Rasch measurement and confirmatory factor analysis were employed. The second goal was to determine if cognitive failures might be predicted with personality factors, general cognitive ability, and the need for cognition. A sample of 552 USAF airmen responded to the Broadbent Cognitive Failures Questionnaire (CFQ), a Big-Five personality inventory, the Reasoning Test, the Speeded Cognitive Ability Test, and the Need for Cognition survey. Both Rasch modeling and confirmatory factor analysis indicated that a single factor dominated CFQ responses. Regression analysis showed that CFQ responses were predicted well by personality factors (R = .60).
Uploads
Papers by Bill Tirre